First off, you don't have a custom process for every random bit of packaging out there, even those that occur as frequently as pringles cans. Second, even if you did, you'd have a massive sorting problem on your hand and those custom lines to deal with each and every products packaging would require an awful lot of space, energy and other resources to cope with.

So as nice as this idea sounds it is not really workable in practice.

The way it does work is indeed, shredding, then float tanks to separate the lighter materials from the heavier ones, then some more stepwise improvements (for instance: to separate out the steel from other metals) and finally compaction and what comes out the other end gets passed on to companies willing to pay for it, and if there is no market, which get paid to deal with the resulting sludge/scraps/goo.

Recycling is not nearly as orderly a process as manufacturing is, and there will always be a residue that simply can not be dealt with economically. Properly designed packaging takes that into account at the time of manufacture to ensure that the residue is as small a fraction as possible.

This is a hard problem, and in the longer term, next to climate change one of the hardest ones that we will need to tackle. The good news is that we could start today.

The resulting sludge is in many cases much richer in valuable (but dangerous) heavy metals than the ores we mine the heavy metals from, as well as a few lighter elements like lithium. Maybe the way to handle this is to split it up into one-tonne or ten-tonne chunks, seal each one in a few millimeters of polypropylene, and bury them in carefully recorded secret locations: in abandoned underground mines, quarries, wilderness areas, on the sea floor. Then, you can sell the coordinates of these secret locations to would-be miners—maybe not in 02021 but in 02061 or 02121. Seeking investors with long time horizons!

More realistically, there isn't really any risk of running out of landfill space for product packaging at anything similar to current consumption levels. If a person ate a 50-gram can of Pringles and a 30-gram bottle of Coke every day, they'd have 29 kg of packaging at the end of the year, or 29 liters, which compact down to a 40-cm-diameter sphere.

I used to periodically visit an ecovillage that handled their (much smaller than normal) packaging waste in this way: they would tamp it into two-liter Coke bottles with a piece of rebar as a tamper, and when the bottle was full, they would cap it, plaster it over with adobe, and use it as a construction brick. A 6 m × 18 m house with 300-mm-thick walls 3 m tall contains 43 m³ of wall volume which can be mostly filled with this kind of stuff: 150 person-junk-food-years of packaging.

8 billion people doing this would produce 23 million cubic meters of packaging per year, which sounds like a lot, but it's an 800-meter-diameter sphere. Lake Superior is 12000000 million cubic meters, so it would take those 8 billion people half a million years to fill it up with this packaging, if carefully weighted to keep it from floating, of course.

So, I don't think recycling packaging is a particularly bad problem. If by "in the longer term" you mean over the next hundred million years, I do agree that we'll need to solve it. But I don't think it's a particularly difficult problem at that timescale. For the next few million years, we have plenty of space to just store the stuff until recycling it is profitable.

It is indeed a hard problem. It's also why manufacturers are the only ones who can realistically deal with the problem. Let the Pringles company collect all the used Pringles cans from the same supermarkets they deliver Pringles to. They have the scale to justify building a dedicated facility for recycling Pringles cans.

Think you're making a lot of leaps here, unless you're very knowledgeable of how recycling works in practice...

Or just don't buy Pringles.