Not sure if I'm answering your question correctly as I do not understand exactly what you mean by splitting reverbs but I'll take a gambit 
I do things a bit differently. If it's a complex track, I have 2 send channels with reverbs, both with a graphical EQ and have one send for low range reverb and one send for hi frequencies.
This means if it's the breakdown and there is no low end I will use the one with the low enchancement, if it's mid track and there isn't much space i will use the one with the hi frequencies and less low end.
Don't know how others really go about it but that's how it's worked best for me. Others do it differently, automating the reverb as the track progresses and making use of only 1-2 reverb plugins in the whole track.