When you are adding in multiple stereo sources eg 6 for example there are some tactics that work.  One approach is that each stereo pair only occupies say a 30 degree chunk of its width and the 6 of them are placed around the L R 180 degree pan space like 6 mono sources that are around 30 degree wide though each.
 
The parts near the middle are best for more important things that are there more often. The wider parts for more sparsely placed material.  You can still enjoy a pretty wide image this way with multiple stereo sources.
 
It also depends on how they step on each other playing wise in the arrangement.  Parts that interweave and don’t play at the same time can enjoy a wider stereo return.  Layers are good with narrower widths placed in 3 or 4 sections across the full L R 180 degree width. Having full panning of both L and R sides is important to experiment.  Automating things around slowly can add a sublime movement to the whole soundscape.
 

codamedia

Anderton
There's very little I don't record as stereo tracks these days, but then again, I use Channel Tools to pan the right and left independently so I'm not limited to a Balance control.

I tend to lean the opposite way and record everything in mono. If it's a stereo part (ie: synth, piano with two mics, acoustic with two mics, overheads, etc...) they get recorded on two separated mono tracks labeled left/right. 100% control of panning without the need of channel tools. If I want to share EQ and/or FX I'll just buss them.
 
It's just a work flow I am more comfortable with but next year I may feel differently

I prefer stereo tracks for all non-true mono sources and use plugs like WAVES S1 to place them where I want them and get them as narrow as I need them. 
 
Using double-mono for stereo as described here IMHO causes too much overhead in terms of routing and over complicates the setup (becomes error-prone) i.e. I want to EQ the stereo guitar track and not a guitar buss (or aux track) that feeds another bus (with all guitars) that feeds master bus etc.

And there you have it...there is no one right way.
 
Dan

i've often use a 2.1 setup (not using LFE channel though) on a surround master buss to allow me a lot of flexibility for setting width on stereo tracks (which may start as mono but using stereo effects etc like modulation etc) as well as placement. allows fine tuning of the overall delay and reverb busses as well...

glennstanton
i've often use a 2.1 setup (not using LFE channel though) on a surround master buss to allow me a lot of flexibility for setting width on stereo tracks (which may start as mono but using stereo effects etc like modulation etc) as well as placement. allows fine tuning of the overall delay and reverb busses as well...

now this is an interesting idea. never touched the surround options. never thought of anything like that ... what are the drawbacks???

bitflipper
Count me in the mono-by-default camp. Reserve stereo for things that truly need it, such as recording acoustical and vocal ensembles. I don't like to have more than 3 stereo tracks in a project because contrary to intuition, too many stereo tracks can make the full mix sound LESS stereophonic, not more.

I agree totally...If the degree of separation in each stereo source is not the same the end result may be surprisingly narrow.

Then you become creative in making them sound wide again. Three is the minimum number of anything you need to create something seemingly complex. eg three narrow stereo sources can be panned in a way as to make the end result very wide.  One possible approach here is the first stereo source panned say 8 o clock 12 o clock the second 10 o clock 2 o clock and the third 12 o clock and 4 o clock.  The more different the parts are, the wider things will seem.
 
Also even when you take panning synths such as this and use something like panagement:
 
https://www.auburnsounds.com/products/Panagement.html
 
In conjunction with a reverb the forward back option steps in giving the centre panned stereo image even more depth.
 
There are many processes than can really widen an already stereo synth patch too.  Like 3 pads for example being layered.  If some of the pads are not that wide to begin with, then you can take steps to create a much wider image.  I got good at this working with some early polyphonic super analog synths of the past that had only one mono output.  I developed techniques for turning those sounds into wide stereo patches instead. When you get good at this widening an already even slightly stereo image is easy.  There are also some great plugins around that really specializse in it.
 

Rob[atSound-Rehab]

glennstanton
i've often use a 2.1 setup...

now this is an interesting idea. never touched the surround options. never thought of anything like that ... what are the drawbacks???

no drawbacks per se but does require more attention to detail since it's not going to simply be LRC given you can set width, depth, and angle (including outside of 90 deg or 120 deg etc) - so watching phase esp for outside placement is important - as always, mix initially with mono monitoring :-)