"It’s also possible that there is such a pent-up demand for housing in San Francisco that no amount of building could realistically do enough to make things truly affordable."
I am an SFBA (Marin) resident with a fairly large financial and personal interest in the health and welfare of the city of San Francisco - so this is something I think a lot about.
I think your point is interesting and should be considered.
Obviously it isn't literally true - clearly there would be a point at which enough housing could be built in San Francisco to hit any arbitrary affordability number.
However, it's worth considering that any politically and physically feasible[1] amount of development might not satisfy the latent demand, nationwide, for a house in San Francisco. It is not inconceivable that there are literally 2 million people in the US that would like to immediately move to SF if housing prices were in line with, say, Minneapolis or Denver.
I also think there might be a corollary to this with regard to homelessness in the state of California. I have read with great interest about the successful "house the homeless" projects that have been implemented in places like Salt Lake City which appear to suggest that you can solve a cities homelessness problem by building enough units to house that population. But what if there are (essentially, not literally) an unlimited number of people willing to be homeless in California ? What if the Salt Lake program really only works because it gets cold and snowy there and, really, there are only so many people willing to be homeless there and you can, in fact, house them all ?
If either of those are the case, I don't know what a good solution looks like.
[1] Including the time dimension ...