I agree, scbrowne, that speed is important in a war. In fact, I thought the whole purpose of your strategy in the first place was to speed up the attack, and you've stated some of the very good reasons for wanting to do so.
But I feel a strong urge (don't ask me why I just do) to point out that speed isn't everything in a war. You attack to serve some purpose. There's land you want, there's a threat you want to eliminate, there are resources you're after, whatever it is, that's your end goal, not just moving fast. If you're taking the territory with a goal of being able to do something with it, then it's important that you not mess up the territory in the process of taking it. Caring for your true objectives is anything but silly. Going to war for no real purpose other than to move quickly and destroy territory is, IMHO, very silly.
Personally, I use workers instead of settlers. If the rail network is incomplete, they complete it as far forward as they can. Additionally, I hold back my cavalry (and modern armor if I have it) for exploits of breakthroughs. The third movement point makes a big difference. Before my first move, I plan my attack carefully. I try to see which cities I should take first that once under my control lead to other nearby ones, of those which ones will be 2 squares away, which will be 3, which might have to wait until the second turn, etc. Then I count up my assault troops and figure how many I'll need, how many I'll probably lose, and how much movement they'll have to expend to get to their targets, and finally start the advance.
Once in a while I come up against a high-culture city that takes more than 3 movement points to reach and attack, so it would take two turns. In those rare cases, a settler plopping a city on the border would let me get in quicker. But these situations are usually caused by rough terrain, especially mountains, and since you can't put a city in a mountain square, this tactic won't help me there. There might be a few occasions where it would help for my playing style, but not many.
This is not to say it's a bad strategy. It just doesn't fit my playing style. I don't expect everyone to follow my particular doctrine, and the tactic you described will probably work well for a lot of folks whose style is different from mine.
For me, blitzkrieg is all about punching holes first, then moving fast units through those holes. Use slow units to make the holes, and fast units to exploit and wreak havoc behind enemy lines, moving forward faster than he can recover. If plopping a city down helps you advance faster, it's a good thing. If it's not a great boost, it's probably not worth the effort and expense. I think it depends most on how you approach an assault and what your goal is in that assault.