Basically it makes the assumption that the current Internet infrastructure has too many issues to be able to make it more secure, and meet the needs of the current world. They are suggesting that it might be best to just start from scratch.
It sounds really good, but a few things that come to mind.
1. How long will it take to get it built and functioning?
2. Who will be the one managing the architecture?
3. Maybe more important, who is going to be funding it, and what kind of controls are they expecting to have?
I think it's good that this is being thought of, and even if it doesn't come to fruition, it will provoke discussion on where the Internet is heading and what's the next step for it.