I don't see the LCR concept about collapsing stereo signals to mono and panning them either to L C or R.
You're absolutely right, assuming the stereo file is really stereophonic.
But if the synth patch is actually monophonic within a ROMpler's ROM, and a chorus effect has been used to imbue it with a faked stereo-ish sound, then the L/R differences are only the result of comb filtering and most of the signal below 500Hz remains smack down the center. This gives you a lot of mud in the middle that'll step on your vocal.
OTOH, if the patch is really stereo, e.g. a mixed orchestral patch with instruments panned across the classical stage, then you've got different instruments on the left than the right, not just separated by frequency. You'd lose something for sure if you collapsed them to mono. The effect would be as if the orchestra had all moved to one side of the stage.
Another example is a Leslie. Much of the Leslie's effect is a moving acoustical phenomenon. A Leslie miked mono has a very different sound than one miked in stereo. However, if you situate a stereo recording of it in the center, it sounds as though the speaker is in the center of the stage. Low frequencies remain in the middle. A real band on a real stage would have the Leslie to one side of the stage. So you have two choices: fold to mono and pan, or use something like Channel Tools to pan one of the stereo channels to the side. Either way you shift the low frequencies out of the center and away from the vocal.