Very nice flushing out the detail of meta models. I actually wrote something about what you call F3 for Deep Learning back in March ( https://medium.com/intuitionmachine/machines-that-search-for-deep-learning-architectures-c88ae0afb6c8 ).
Now this idea of FΩ, well that’s entirely different observation from what I wrote and look forward to the next post.
I don’t think there’s a higher level than F3. That’s because if F3 folds back into itself, you get, likely FΩ.