Radiation hormesis is not "fringe science". As only one example, consider
http://www.jpands.org/vol13no3/cohen.pdf which is an overview from 2008 of some of the issues raised against the linear no-threshold dose model. You'll note that it was not published in some 'quack journal', and draws its references from many other non-quack entities.
In addition you're assuming that changes in regulatory setpoints are purely based on proven safety effects, but that's not the case here. Indeed, regulatory changes happen frequently that are not based on safety (e.g. yellow light timings).
What is true is that if you assume, as most health physics societies and regulatory organization do, that radiation exposure should be kept ALARA ("as low as reasonably achievable") then you should reduce regulatory thresholds for no better reason than that it becomes possible to do at all.
In other words even if you find a 'safe' level below which you can't appreciably detect an increase in cancer risk, you'd still set the regulatory threshold far below that if you could (and keep dropping it as technology improves), because regulatory commissions are operating under the assumption that any exposure to radiation represents some additional risk of later cancer.
But it's important to keep in mind that it is an assumption, a convenient and conservative model that is used for ease of regulation, work planning, etc.
That doesn't mean the model is accurate, and the actual biological response to radiation of different types and exposures continues to remain debated at low levels of exposure.
After all, if what you claim is true then frequent flyers, air crew, and pilots should be dying at a ponderously higher rate of cancer incidence, as should residents of high-altitude areas who are exposed to higher cosmic ray flux, as should residents of Ramsar Iran who are exposed to high levels of background radiation from uranics in the soil and water.
But none of these people seem to be as effected by radiation as they should be. The question is why and yes, it's still a question.
It sounds like you're at least interested in the topic so I'd highly recommending researching further. For instance if you do so you would find out why mutation in a single cell is not at all the crisis you think it is, or even that surprising, because you would know how the body handles such cells (which occur all the time from multiple different sources).