I'm new but I had at least what I think is a new idea.
The idea was how to really take advantage of the fluid reactors and I hadn't seen anyone do it yet.
I was curious what the theoretical maximum for energy generation was, which also needed the maximum thermal transfer rate, and found that you can actually run a reactor with a 4 by 4 square of quad cells continuously in a standard reactor. Cooling was done by transferring the heated components into a fluid reactor (or in my case 10 of them) where then of course that could be turned into EU. The pictures are from my reactor I have running on my server (though only 3*4 quad cells).
For this setup I made use of Enderio to transfer the heated components around.
Though the way I have it running in the picture is not the full 4 by 4 square it still puts out a lots of power as there's the ~5000 EU from the reactor directly and another ~4500 EU generate from the heat. The full one I had running for a bit in creative maxed out with around ~8000 EU from the standard reactor and another 8160 EU generated from the heat. Until it exploded... (thus the slight down size in the picture) The picture below shows the full 4 by 4 square of mox cells that put out around 6000-7000 EU directly plus another ~7500 EU generated from the heat. It required some heat plating to be stable else the temperature had rather large fluctuations.
The reactor heat exchangers though if you're thinking about this though don't add up with their only 72 heat/second as there's only room for 38 of them (with the 16 quad cells) 38*72=2736 which is far short of the 5440 heat/second that the cells would generate. However the heat exchangers adjacent to the fuel cells get the heat from those cells directly transferred to them increasing the effective heat transfer rate. So effectively the heat transfer rate of the reactor exchangers only need to cool the middle quad cells not directly adjacent to any heat exchangers.
Update: I've made some tweaks and managed to get the full 4 by 4 square of quad mox cells to run stably, seems my math was a little off and I was able to include some heat plating which dropped the temperature making it more stable. I've updated the picture here to represent that. I did add a 11th fluid reactor though 10 as should work it was a little too close for comfort.