I'm not sure what you mean by "over-simplified non-linear models", please elaborate.
I would say in general though that the problem is not the models but the parameters - this is such a new virus that modellers have to make various assumptions about things that we can't possibly know yet.
(Genuinely asking, since I wasn't aware epidemiology wasn't a science.)
No lockdown => hospitals overwhelmed => people can't get care they need or people avoid hospitals to avoid getting sick => excess deaths
Lockdown => hospitals less overwhelmed => people avoid hospitals due to lockdown or to avoid getting sick => excess deaths (but less than the no lockdown scenario)
Here in Austin our medical community is practically begging us to take better precautions because our hospital system is starting to hit capacity due to the second wave we're staring down [1]. Saying it's not 'accepted' in the medical community seems wrong.
>Lockdown => hospitals less overwhelmed => people avoid hospitals due to lockdown or to avoid getting sick => excess deaths (but less than the no lockdown scenario)
Right, but eventually as things taper off there's a crossover point where the first option will have less deaths going forward. Where that crossover point is the point of contention.
Which then means you have to go into lockdown again while people yell at you for going into lockdown not understanding that the alternative was made nonviable due to them not taking basic precautions.
This claim is not accepted by all in the medical or scientific community. This is just the line from the epidemiologists who have the most influence, and it has been amplified by the media into some sort of unquestionable creed. My guess is the damage that these epidemiologists have done is going to be far worse than the disease itself. If you can't see that their over-simplified non-linear models are essentially just guesses too, then you are missing a big part of the picture.