To reduce the reflections created by the window try something on the wall opposite that adds difraction so the sound doesn't keep bouncing from one to the other. A "live end/dead end" approach might be appropriate, where one side of the room, typically behind the listening spot, is left acoustically "live" and the opposite end of the room gets the damping. It's very much a matter of experimentation though.

One thing that might help quite a bit is getting the monitors to four or so feet from you and the same distance apart Nearfields are generally intended to be used close to the listener, and often really need to be at least a couple of feet from a wall either to their side or rear, otherwise the bass gets ramped up. Though that's more of an issue with larger speakers than 5". Setting up a position that works equally well across a wider listening area can be very difficult indeed. Speakers too far apart don't produce a wider "sweet spot" but a hole in the middle which makes hearing what's going on in the centre very difficult - which is one reason most live mixes are in mono. I'd be tempted to put your current nearfields close to the listening spot and a couple of 8" or so monitors further away, though if it's a choice between more speakers and killing obvious room reflection and reverb issues I'd go for sorting the room out a it first.

There's no need to spend huge amounts of money on commercially-made bass traps or other reflection-killing traps, much can be done using panels made of heavy grades of rockwool, especially if it's spaced a few inches off the wall so sound passes through it twice before it reaches the listening position. Traps place across corners often work better for taming bass than ones on the wall. The ceiling is also a good place to put traps, and probably isn't cluttered with other stuff. The door can be treated as well.

Bear in mind that many commercial audio tiles really only reduce high-freqency reverberations. A 2" thick acoustic tile might only be working in the 10KHz+ region. And diffraction is at least as useful as trapping and can massively reduce standing waves and flutter echoes.

Furniture and bookshelves make good absorbers/difractors.

This really isn't something that is easily answered on a forum because it's complicated. Try searching the web for "diy acoustic room treatment" and similar phrases.

This is why it's critical to get near-fields as close as possible.. the closer they are the less the room sound will be involved in your listening experience. So if the room is not a proper studio set up we strive to work real close to our monitors. I think it's a waste of money to ad lib a room treatment.  All that will happen is you will "change" the room. you still won't have it right. 
So just get those speakers close to you. I would think stands that can be easily moved when guests are present would be ideal. It may involve using a securement so they don't fall off when moved. 

Cactus Music
 
I think it's a waste of money to ad lib a room treatment.  

Say what who now ?

Here's an update.  I got the stands in place and both speakers are symmetrical, 4 feet from my ears and several feet from the nearest wall.  That is sounding good.
 
This allowed me to hear what was undoubtedly an ongoing problem.  My first interest was to hear how stereo separation improved.  I played some music back though Media Player and I couldn't hear any separation at all.  I went into SONAR and created test patterns where I could manually pan for certain.  That worked, but I wasn't hearing the separation in Media Player.
 
It turns out that on my audio interface (Presonus VSL1818), the Media Player feeds a different par of ports than SONAR feeds, and I had left the default setting to sum to mono.  This is maddening because this problem has been happening for most of a year, and I only discovered it today.  I don't use the Media Player that often, but it never would have been playing correctly since I got that audio interface.  When the speakers were across the room, I guess I didn't expect the stereo separation to be that apparent.
 
And I think that when I rendered directly from Finale, it was going through those ports playing mono.  I always figured Finale just wasn't capable of a very good job of separation and never really investigated.
 
Another thing I discovered is that my two speakers have vastly different output levels.  They both seem to be putting out a good quality sound, but the right one is half as loud as the left.  This may be a wiring problem, but I was able to adjust levels on the speakers to compensate for now.
 
So I guess this is a confession that I wasn't hearing things very clearly.  Definitely a big improvement now.

My comment is this. If you haven't done something silly with audio at some point then you haven't done anything with audio at all.

Great news.
 
Unbalanced powered speakers are not unusual. It's why they have individual volume controls on them. But it could also be an acoustical phenomenon. Either way, just run a mono signal through them and adjust so it sounds centered from the mix position. You may have to revisit it after installing absorption.
 
My previous room also had large windows on two sides, and they definitely had a strong influence, especially when using a microphone in the room. But because that room was dedicated to recording and mixing, no longer serving its previous purpose as a guest room, I took the extreme step of covering the windows entirely with rigid fiberglass. I left the blinds in place to hold the fiberglass in place but left them in the open position. Yes, I did lose the calming effect of looking out at trees, but made up for it with mood lighting.

John
My comment is this. If you haven't done something silly with audio at some point then you haven't done anything with audio at all.

Best comment ive seen today.
that little saying can also be used in a lot of different ways....as in replace the word "Audio" and insert the word "Life"

Run a 20 Hz - 20 kHz sweep tone through the speakers, and determine if you hear huge frequency response variations. Some EQ in the master bus can take out some of the worst variations.
 
As to speaker level mismatches - this can happen if one is being fed a balanced input and the other, an unbalanced input. Check your cables and connections.

As to speaker level mismatches - this can happen if one is being fed a balanced input and the other, an unbalanced input. Check your cables and connections.
 
Absolutely!! The minute you said this I went,, oh oh, bad connection somewhere. 
Or the audio interface is not outputting evenly Or the speaker is damaged. 
Run a diagnostic swapping. 
It helps if both cables are the same but it should not matter if they are both balanced and in good condition. 
I'm anal about cables so mine have to match 100%. 
 
Swap the cables at the interface. 
----If the the quiet speaker is now louder then it's the interface or the cable. 
      If it stays quiet then it's the speaker or the cable. Not the interface. 
Put the cables back where they were then:
Swap the cables at the speaker
----If the quiet speaker is now louder then its the cable.  
----If it stays quiet then it's the speaker. 
 
 If you rule out the Interface and the cables and one speaker being radically different in volume at the same settings,, I would be a little worried that it is damaged. I can see a small difference is possible but Yamaha is not a brand I can see having this issue unless it's damaged. 

Anderton
As to speaker level mismatches - this can happen if one is being fed a balanced input and the other, an unbalanced input.

It is all unbalanced, and the weaker speaker has a cable that is about 10 feet longer than the other. 
 
It is unbalanced because I run everything into a mixer, then feed it through a Samson C-que8
http://www.samsontech.com/samson/products/processors/c-class/cque8/
 
which is a cheap way to manage several sets of speakers.  And the only outputs on that box are stereo 1/4" headphone jacks.  With the new speaker placement, the speakers are only 5' from the C-Que8, so I expect if I get 2 6' cords (and my stereo splitter cord), the levels will be very close.
 
I should probably invest the time to rewire everything.  I think the Presonus VSL1818 has enough inputs and outputs (plus its own internal mixer) so that I really don't need an external mixer or the C-Que8.  That would probably clean up some low level noise I have always put up with.  The noise is only in the monitors, not on the recordings.