>...slow the flow. more contact time... more removal of DOC's
Very good. See, Ben, there is a direct, mathematic relationship between flow rate and contact time in a recirculation scheme with a specific pump requirement. That is, you can'y increase one (flow or time) without using a larger pump. The relationship between contact time and DOC removal is a bit more complicated, but for this discussion, it is suffice to say that the more contact time, more DOC removal.
Why do I say this? This is because you are right about your statement. But then, let's think about why these recirculation advantages is used in other devices that we use regularly in our hobby. The move obvious one is calcium reactor (the type that use CO2). The same advantages hold. More contact time, more dissolution of CO2 and cal/carbonate. But why calcium reactor does not brag about slow flow? That's because it is more efficient to recirculate (nothing is free, and the price you pay is slower flow) within the reactor so that the *net* amount of calcium carbonate can be dissolved within the same amount of time.
I won't tell people to use a one-pass reactor inside a sump and expect the same performance as a recirculating calcium reactor. This is because the sump is open to air, and much of the CO2 will dissipate if such arrangement is used.
But it will work for a regular skimmer to be a recirculating type by the use of a sump. The fact that the saltwater temporarily exposed to room air does not decrease the foam efficiency.
Again, there is no free lunch here. Recirculating decrease the *net* flow rate and increase the contact time, both of which can be accomplish with a single pass skimmer in a container (sump) that pump water up to the display tank (or other chamber of the sump) slowly.