OK.  I've taken the time this year to slow down on writing and recording and actually LEARN some of the finer processes.  I've finally sat down and actually STUDIED equalization and compression and how they actually work where before I just added these things and either used presets or turned knobs blindly in hopes that it would eventually make things sound better.  When I look back on my settings on the projects that I did before my enlightened education I have to laugh sometimes when I see things like a compressor setting with the threshold way open.  Hell, you might as well just delete the damned thing because at +15 db YOU'RE NOT COMPRESSING!!
 
So, I started researching parallel compression.  I ask you to verify my interpretation of it against what it really is because some of the YouTube videos (you know, the "Hey guy's!  What's happening?!  I'm gonna show you how to...........) can be very vague because they either use terminology specific to their DAW (and none of them EVER seem to use Cakewalk/Sonar!  grrrrrr!!) or they just explain things at such an advanced level that newbies or intermediates can't pick it up.
 
So, please correct me if my interpretation of how to apply parallel compression is correct or not.  Here goes:
Parallel compression is done (let's say on a vocal track) by opening another channel with the same vocal track as the original  (so you have two channels with the same vocal track).  On the first, you process it to taste (iow, get it to sound as good as you can get it) and the second copy you compress heavily and bring it in behind the first and that should make the overall vocal sound somehow stronger or better.  This is how I've been doing it and it has done nothing to improve the sound except to make it louder.  So, am I doing it right or not?  Apparently not, because it's not improving anything for me, so I figure I'm just not doing it right.  Please educate me on what details I'm lacking.  And while you're at it, if you can, send me a link to the best education on side chaining because that's the next endeavor I'll be working on.
 
Thanks so much!!
 
Your brother in music;
John B.
 

Hi. I usually set up a send from the track I want to parallel compress, to an extra/parallel bus.
There I compress a bit extra and blend in to taste.
Others might do it differently.
All the best.

Zargg
Hi. I usually set up a send from the track I want to parallel compress, to an extra/parallel bus.
There I compress a bit extra and blend in to taste.
Others might do it differently.
All the best.

Ken
Thank you for your feedback.  I do realize that many people do it differently.  I mainly wanted to know if I have the general concept correct, but I'll try the Send/Return thing that you do.  I remember that one of the YouTubers said that's how they work too.
 
JB

I do the same as ken does, no need for duplicate tracks, it’s called send + blend.
Feed a send off of the tracks you wish to parallel to a bus with a compressor and control it using the bus fader

You should not have to do anything.  It depends on the compressor itself.  Maybe Cakewalk compressors don't have the facility built in but Studio One compressors do and most third party compressors also. There should be a mix control on the compressor itself. Set at 100% wet you are hearing the compressed sound only. As you bring the mix control back from that e.g. half way or 50% then you are hearing both the compressed signal and the input or  uncompressed signal at the same time.
 
The other approaches mentioned here are good if you want to do some other processing to the two signal paths that may be different but in general you won't and the mix control will do it all for you instead.

If I understand it good, the point of parallel compression is to get a better result with a compromise. You get the dynamics of the original signal + the dynamics of the compressed signal. Which in sum gives a louder, more even signal ... but still with some of the original dynamics. 
 
So, I think I would first deal with shaping the dynamics of the signal, and then apply effects (EQ) to it. I think this will leave you "free'er" to change the dynamics of the signal at a later stage, during mixing, without also having to rework the effects part of it.

Many of the more modern Cake comps in the PC have a blend feature, or parallel comp.  the incoming signal is split and the comp works on one stream and the other is clean.  Blend to how you like it.
 
And you have the right idea about putting the comped signal "behind." Often the compressed signal is squashed like a tick - to bring out all the nice things about heavy compression.  Even the volume and maybe put a spitty edge on the signal so that it sticks out.  Add a little of that squashed signal to beef up the main cleanish and dynamic signal.  Just like finding the problem frequency w/ EQ - exaggerate the frequency gain and sweep the frequency to find which one sticks out and is the problem.  Exaggerate the compression and then blend it back into the mix until it stabilizes the main track.

The method that's usually described is actually just one of several paths to parallel compression. You can, for example, mix the dry signal under the compressed signal rather than the other way around, which is a more profound difference than you might initially think. Or compress a band-limited version or a distorted version. Or exclude certain instruments from a parallel drum bus, or alter the send levels from different instruments. 
 
The important thing is to keep in mind what parallel compression actually does: in the end, it's upward compression. Meaning that quieter sounds below the threshold get louder, such as cymbal tails and bass drum resonance. Having that mental picture will guide you in how to apply the effect, and inspire creative variations to it.

Parallel processing in general.. is an old trick.  Guitarists who use 2 amps at 1x are generally doing the same thing.  1 dirty 1 clean.  
 
Don't get confused about the technical things, you'll get used to it the more you do it.  You will eventually see that there are only a few things that actually do happen in the world of mixing, and different manufacturers simply have different names for them.  As you uncover new possibilities, jot things down in a notepad on your desk.  1) for future reference, 2) writing down things will help you remember them as well.
 
In the recording world, its best to stay away from things such as eq on your compressed side, as these can cause some phasing issues.  I've seen some great things in that world.  Check this out:
 
https://www.youtube.com/watch?v=ajSv9Wj81tc
 
As far as sidechaining goes, heres one I made awhile back.
 
https://www.youtube.com/watch?v=WJoruTWCnyk
 
And another
https://www.youtube.com/watch?v=glRU9H37Fbo&t=2s