Exactly Sula, I am confident we can go significantly further than 100nm as there is no reason not to be able to and then GT just stated it which puts any doubt to rest.
It will likely be the same thing as with Silicon and there being that threshold where it scales down to a certain point, be it 15nm or 10nm or further and the cost-effectiveness of scaling down any further just doesn't make sense.
I also believe that it's something that will be governed by the mitary applications. Similar to how F1 works for the car industry, if PTK continues its ties and work for military applications, this will likely be the driving factor for driving down size, increasing efficiency and any other characteristics as quickly as possible with the consumer market lagging behind a few years.
Also using common sense from the business side of things, if we're already significantly, and I mean significantly better than the current best silicon chip than the industry will want to milk it like it did with silicon in order to make as much money as possible. This will likely be given a longish timeline of lets just guess for fun 10-20 years and then whoever aligns with PTK will be making gravy until the next solution to the innvitable wall comes along.