Push Pull generally increases air flow by 10-30% (21% average) ... and oddly enough, ytesting at martins liquid lab showed that having slightly different speeds / Sp actually impo0rved performance ... which was a result one wouldn't expect,
I have got assume that you bought the H7 because of a tight fit situation... are you sure you could mount a 2nd fan ?
From your original post, I thought you were talking about a fan swap. Here we see the following for the UT60:
1400 rpm gave 210 watts of cooling
1800 rpm gave 269 watts of cooling
2200 rpm gave 327 watts of cooling
Averaging the 1st to will give us an approximation at 1600 rpm, say 240 watts. At the 2200 rpm, we see a 35% improvement which will cost a significant boost in noise. However an air cooler behaves differently, I still would expect the boost to be about 25% or better.
If you can use push pull, that is... it does fit, I would not be concerned about the different speeds. Adding a fan, however, tends to have a lesser effect. Here we see temps drop from 52.75 w/ 1 fan to 51.75 w/ 2 fans to 50.75 w/ 3 fans
http://www.vortez.net/articles_pages/phanteks_ph_tc14pe_cpu_cooler_review,14.html
If you are looking for more performance, and are willing to settle for more noise, than the faster fan should get you some of that. As opposed to "throwing it away", I would put on the old fan (if it fits) and see how it works. Martin I think surprised us and his readers with his test results but there is no way to judge how wide this effect might be or what % of possible combinations would benefit. But again, it costs you nothing but time to see if you can realize the benefit that his test showed. Logic dictates to have the slower fan in push and the stronger fan in pull.
If you go this route, please post the results .. I musta had that silly puppy dog tilted head quizzical look on my face when I read Martins article and have been occasionally tossing it around in my head ever since.