Well...you can apply comps and EQ to individual tracks too...not just the whole mix....and it may be more advantageous to do that to individual tracks rather than the whole mix, because you can be more selective what type of comp or EQ you are using on the guitar track VS the piano...etc.
Then if you need/want some final, overall adjustment to the mix...you only add a pinch of compression and EQ to the whole mix, rather than having to hit it harder...which rarely works well, as it becomes a tug of war. You tweak it so the guitars sound great in the mix...and all of a sudden the vocals are suffering...etc.
While not a *rule*, because as I said, an Aux is just a signal split that allows you to do "something" to it and then you add it back in to the mix...but in general, Auxes are good for FX...but *processing*, like compression and EQ is best done in-line with the signal rather than in parallel-n-combine fashion, though there is a technique where you take a split off a signal...compress the crap out of it, and then blend it to-taste, back in with the uncompressed signal. It's been used a lot, and it's often called "NY compression" or some such thing (aka parallel compression). It lets you keep the body of the signal untouched, while still having the sound of an over-compressed signal too. It was mainly/originally done with drum tracks...I beleive.
Of course, you can also NOT over-compress the split signal...just do as much you like...but that's how the technique originated.
Mind you...I mix mostly OTB....but in the DAW, you may have even more complicated/involved Aux buss capability than what is on a basic hardware mixer.
