Proper FILL should be mono, bandpassed, delayed and attenuated. It should be simulating the reflected sound from the rear of the venue. The reflection will kill the highs and lows and will decrease the level compared to the direct sound. It will also eliminate the directionality that requires stereo to reproduce. You can actually use a single driver to get effective rear fill.
My opinion is that unless you listen to a lot of recordings of live performances, rear fill is not needed or wanted for accurate reproduction of the recording.