My front fans are about 2cm back from the front filter (see this). However, the front filter is surface mounted the the face plate. The fans themselves are part of the case proper. Because the fans have square fans, there is no recirculation (leaving one of the fan spaces unoccupied might invite a limited amount of recirculation; but both spaces are occupied). Basically, as covered in my review of the CM N600, the filter has 50% more area than the fan surfaces. It and the faceplate create an air chamber.
And the proof is in the pudding: the filters accumulate dust that must be cleaned periodically. I can measure the outflow from the front fans with an anemometer. That outflow is much higher with relatively high static pressure fans than with relatively high cfm fans.
I don't care if aerodynamic theory says static fans can't pull against a filter. If theory doesn't explain the data, you must chuck out the theory and get a new one. Countless people have discovered that fans with high static pressure do quite as good a job in pull as with push. BTW -- the same is true for high density rads. In fact, fans work better in pull, and high static pressure fans work best of all.
I want my case and my hard drives cooled. I have both relatively high CFM fans and relatively high static pressure fans.
I say "relatively" because I use quiet fans. But I can and do measure both noise and airflow. I can, for example, choose an NF-P12 on ULNA over a GT, either AP-12 or AP-14 on 5v (relatively high static pressure fans) over an NF-S12A for behind-the-filer work. I can also pick the low static pressure NF-S12A for my mid-case fan.
I can't comment on AF and AP fans. I was done with fan-buying before they came out, and I have received none for review.