You're using two reactor's, one for GFO and one for GAC? Would the BRS dual reactor be able to do the same thing but with only one pump?
You can, and many folks do. But, I think the best case scenario, and the one I preferred to go with is the two reactor route, so I could better tune the flow to each reactor. Contact time with the water and media is important, but the GFO needs to be tumbling a little or it will clump together and solidify. The GAC should be compacted enough so that it doesn't tumble, so there should be different flow rates for each media. The other issue is flushing the media when you first install it. So, unless you rinse the media, before you install it in the reactor's cartridge, which I think would be a PITA, whatever media you've placed in the first chamber, typically the GAC, is going to flush into the media of the second chamber. With the individual reactor's, I can change the media, rotate the discharge tube into a small bucket, and run the pump for a few seconds to remove any remaining fines, and then redirect it back into the sump.
I installed mine, by hanging them on the side of my sump, but I've seen a few that mounted them above/near the sump, and just have the pumps and discharge tubes running into the sump.
You could run both the individual reactor's on a single pump, but you would have to get an additional tee, use a large enough pump and utilize the ball valves to regulate the flow to the individual reactor's.
You could also run these by mixing the media's together, about 2:1 GAC:GFO, and compact them enough so they didn't tumble at all, since the GFO would grind the GAC into dust. But you may not get the full value out of the GFO, since you'll have to replace it when you replace the GAC, in which case I would use the basic GFO, instread of the HC GFO.
Another possibility, if your system is small enough, is to run both in a single reactor, by separating the media with a sponge.
So there are a couple of options, but I think the best is to have a reactor for each media.
How big is your system? If it's small enough, maybe one of the single reactor solutions would be best.