Just to get an idea of the size of this track: It's about 70 tracks deep + about another 30 bus channels on top of that. It's a mess. But it's how I work.

I've held off using compression on the track until now because the tune was a collab and I didn't really know what the other guy was doing with it when he had it.
So anyways, here's how I'm thinking about doing it myself:
I'm going to go into the arrange and label everything properly (drm for drums ect)
Then I'm going to set all the volume faders back to 0 and bounce everything as audio tracks to a clean new arrange. (Here's where I need a bit of help in making a decision. I have a lot of tracks where I've used stereo imaging. Do I bounce the tracks wet with the imaging on? Or do I bounce them dry and lose the same spread that I've got them in right now? Same question goes for stuff like panning, delays, reverb too.)
From here I'll give everything a final once over individually and tweak the volume, adding eq and compression if I think any certain track needs it
Then I'll group them into bus channels. One for drums, bass, synths ect. Again adding eq and compression needed.
On my master channel...I just put a low pass filter at a really low setting (25hz) and a limiter so the bounced audio isn't too low in volume when I put it on Soundcloud ect. (This time I'm considering other options on the master, that's if anyone has any good ideas)
So yea....if you think I'm missing something, or you've got a better way/idea....please tell me.

Thanks in advance.