So, more light actually turned it dark.
I think it's important to note that you still had a lot of green growth. The "black, oily" growth that has been referred to in the past as the indicator of high nutrients, needs more frequent cleaning to avoid shading of algae underneath it, etc, may not be what you are seeing here.
When I first fired up my UAS test box, I got a lot of dark stringy growth - like dinoflagellates (SW). I noticed that once I got a base of growth (green) going, if I increased the photoperiod, there was a point where the dino growth took off, and if I lowered it beneath that point, it was not as much of an issue. Still there, but not an issue.
I would agree that you want to clean off the dark growth, but if you're still getting the green growth, then it's not nearly as much of a concern - just let it be. I do. I still get the dark slimy growth on some of my screens. I think there is a relationship between how much of this growth you get and how much flow and light you have.
You may have had the right balance of flow and light with the LEDs and diffusers, so you didn't get the dark growth, but you didn't get enough of the green to filter to your liking.
Add more light by removing the diffusers and lengthening the photoperiod without increasing flow, and you now set conditions for the dark growth to fill in.
It is just a theory that fits the facts, but this isn't the first time I've seen it - personally. It is however the first time I've seen it happen in FW.
Let's see how the CFLs do though. I am interested in that result as well. But if it's no better, then I would plan on going back to LEDs, no diffusers, and increasing the flow.