"...if you're sending the file to a professional mastering engineer then the level of the file doesn't matter (as long as it's no unreasonably low like peaking at -18 dBFS)."
I need to be told that, David. I can see how I got conditioned by all the talk that says that it has to be between -6 and -3 dBFS. I guess that number is given to make sure people stay on the safe side, but it made me, in case of this mix, unnecessarily spending time fiddling with the levels of each track since I was peaking initially peaking at around -10 dBFS.
As an aside, leaving alone the fact that I know that trial and error is part of any learning process, I know that I spent way too much time adjusting the levels of my tracks in this mix. Why? Because I'm confused about the best practice with that between using a gain plugin on a track, using the gain tool for individual regions, and using automation. Because of that, I ended up with many regions having different settings involving all three, which made for a lot of—I suspect—unnecessary work. I need to learn more about a sound use of these three elements, to make mixing easier.