Ah geez, you've got some 'sperimenting to do. Because the possibilities really are endless.
I think the last aux buss I set up for reverb was a chorus (a thick, lush chorus, 100% wet) feeding a reverb (also 100% wet). That sort of "detached" the original signal from the reverb and gave the verb a very "dreamy" quality. Fed that into a very low ratio (1.2:1?) compressor in the same chain with a very low threshold so the louder the signal going into the verb, the less verb would come out. The louder parts of the song were left without the "crowding" of too much verb, while during the quiet passages, the verb was allowed to invade the space much more.
You can do the same sort of thing automating the reverb sends and skip the compression, but I tend to like the way it "glues the verb to the mix" (for lack of a better term).
But the same thing goes with everything else -- You can squash a signal going IN to a verb, you can squash the signal coming OUT of the verb, or a delay, or a flanger, you can flange the signal going into the verb, coming out of the verb (where you might not be want to be 100% wet on the flange, but it can certainly spice up the verb) -- All things you can't do when the effect is in-line with the source.