You may want/need to consider this in a wider context (see my other post), if this temporary "overfueling" can help you to prevent "knock" under specific load conditions, then you can run much more agressive mappings, and it helps you to accelerate better, reaching (car) speed faster, and carrying more speed for longer.
I think Coopers article is "good", and goes some way to explain the underlying problem to the normal fan. I'm not so convinced that we talk about "ballons" or excessive "storage" here (because there where already regs in place to prevent this), I think it's a lot more subtle then this, but I see what they are aiming for & where the advantage would lay. It's just application of "proper" engineering / system dynamic principles / playing the margins.
The problem is that these are assumed margins. I see it like when I see a speed limit sign on the road. It's an absolute limit. But it's not practical for auuthorities to measure everyone's speed every single millimetre. So I can assume there is a margin there but technically there is not.
That's why I feel that, if the FIA think there is any team that is not strictly adhering to the limits defined in the rules, then they should say that the suspect teams are cheating, regardless of whether it can be proved or not, or whether they intend to take any action if it can be proved.