I've seen that argument made many times, to be frank it's not actually as well founded as it seems.
This is because it confuses two very different things: ease of production of natural resources and cost effectiveness of production of natural resources. In the early stages of industrialization ease of production is paramount, but in later stages quality, production volume, and per unit cost is paramount. Early resources can be very much more expensive to produce and very much lower quality (as they were historically) and yet still be viable. Take coal for example. A small coal mine near the surface can be a quite viable source of coal in early industry, but such coal mines would not be economical in the modern era. Because modern coal is higher quality, much higher volume production, and mined out of areas where it is generally most amenable to industrialized production. There is an old abandoned coal mine within walking distance of me right now, I've been there and picked up coal off the surface of the ground. There's still literally tons of coal there that could be mined quite easily by hand. But compared to other sources of coal in the 21st century it's simply not economical to bother worrying about trying to extract it to depletion.
And the same exact principles apply to essentially every other natural resource. For example, we no longer mine bog iron because it's a very manual labor intensive work compared to, say, shoveling hematite sand into rail cars.
Additionally, this argument ignores the fact that vast amounts of resources will exist in the form of remnants of our present civilization (only the tiniest fraction of which will be vaporized by nuclear explosions). Huge amounts of iron, aluminum, glass, etc. will be left over, short-cutting the normal production process as for pre-industrial processes.
Re-industrialization after a collapse will be different than past industrialization but I see no indications that it will be very much more difficult and certainly not impossible.