As far as new networks go, that appears to be happening.
Has already happened, Nano is feeless, currently supports 250tx/s on main net and requires the energy of one windmill for the entire network.
It's also more secure than Bitcoin due to a higher decentralization coefficient.
It was expected (again, as stated in the whitepaper) that truncating transactions (in order to shrink blocks that are "old enough") would be able to manage block size well enough to stay reasonable.
In my experience (circa ~2011,) this expectation seemed to be generally accepted without much question by anyone talking about bitcoin. If it was acknowledged as a flaw it was usually hand-waved as only likely to be a problem 10+ in the the future.
edit: Or that a new network would learn from the bitcoin experiment and implement a protocol that works better at large scale.