In a roundabout way, I think that what we mean by a diagnosis of ME/CFS is that there are holes in the field of diagnostics, and researchers either don't see them or are willing to pretend there aren't. Diagnoses like ME/CFS ignore that problem by defining a disease with symptom-based criteria, or by denying the sick are sick. Both routes effectively absolve diagnostic entities, and - more importantly - institutions that endorse status quo, of any responsibility.
Mainly they conflict with the main diagnostic process in health care: differential diagnosis. For which there apparently is no plan B, only a plan F, in the biopsychosocial model. Which is the root cause problem, a rigid system of rules where if those rules fail, whimsically evil fantasy takes its place. This is why they fail systematically when dealing with common symptoms of illness. This process is simply not meant to work for problems like this, and they won't change the system or have a plan B.
Pretty much the intent behind my current signature. It's a rule that works a lot of the time, and even resembles how machine learning works, with the same overfitting/underfitting pitfalls. But where this rule fails is the places on the maps where there be dragons, and fairies, and weird ghostly things. Not other places and people like the ones that are known, just not known yet. Nope, instead it's the weirdest mystical crap in the modern world, of the exact type that has been left behind through the scientific process. None of which is reasonable, and so not possible to argue with reason.
And it's probably not even accurate to suggest that it probably works most of the time, because there is literally no way to know. We simply don't know how large the set of all diseases and medical conditions is, or what % of those that are 'known' are actually accurate, rather than close enough to be somewhat useful.
At this point it's almost a philosophical problem, about what is knowable and what isn't, and how to deal with the unknown, or even the unknowable, but a quick look at when physicians attempt philosophy, such as Sharpe's "illness without disease" whine, makes it clear this is a complete dead-end that will never solve this problem. Physicians are trained in exactly the wrong way to do philosophy about their job. They are trained as mechanics, on specific makes and types of car parts, and nothing else. There doesn't seem to be any generalizable knowledge in there, every problem is its own unique snowflake where all other problems before don't apply.
They can handle problems not in the manuals about as well as a mechanic could handle a vehicle produced in an alternative universe from a completely different starting set of designs, technologies, processes and components. Or how well a car mechanic from the 1900s could fix a modern electric car. They have a very fixed event horizon outside of which nothing can occur, since they can't see it and give it reality by the simple act of confirming it.