Let's face it, sleep is a dynamic process not rest so it is not wonder our ME bodies can't do it properly.
Maybe in the early days the body is trying to fix viral damage in the brain by having lots of sleep but eventually there is not enough energy produced so it can't be sustained.
We do not know how hard the body systems are is working to compensate for our low levels of energy. The brain, heart and lungs take what they need as a priority but there may come a time when the flow of oxygen does not leave enough for the extra effort of the brain to fix its damage so it stops trying and maybe does not have enough to spare to initiate proper sleep.
We may still feel we are doing OK but we are working to our limits without the spare energy available to healthy people.
What you suggest sounds pretty viable, and makes me think of when you see on a TV programme someone put into induced coma to heal - now I'd be interested to hear the science and reasons around those scenarios.
It's an energy-limiting condition. It should make far more sense than many of the other pushed ideas, just by logic being employed as long as that categorisation is heeded. These things could be indicators of progression and things going on in the body.
Which makes the 'symptom-focus' approach massively flawed, particularly when instead of hearing the exact description of issue described by the patient (can't sleep when overdone it) into another box simply because it is already there by virtue of one of the words ('sleep'). Before insulin was discovered people weren't sent to fatigue departments and other things based on their broader dumping pot for each symptom were they? You are basically separating out key indicators that lead to what is actually going on in the body then lowest common denominatoring across conditions (rather than even compare and contrast which could be more interesting insight).