Elon's pauses make it clear to me, and presumably to you, that he was checking what she said versus what he was thinking, and finding that, to a layman, the two would seem similar. Even if underlying technologies are different.
His wording suggested strongly to me that his solution was a tube. Did it not suggest that to you?
No matter how many times you say "pointless complexity", that doesn't mean that there really is pointless complexity. (And we're disagreed on the complexity.)
I'm quite aware of how hard it is to push air out of the way. A quick comparison of the mass of the plunger and the mass of the air says that a bubble of air will not actually slow the plunger down much. Being in air constantly would, but a bubble of fixed size wouldn't. (You'd have to think about what effects it might have on passengers inside.)
There is absolutely no need for a tight seal, as long as the flaps open and shut at the right time. You can stand a long ways away from a moving truck and still feel the wind pushing away. And the closer that it gets to Mach 1, the farther away you can feel that gust.
The friction losses are hard for me to estimate. The upper limit on the temperature of the plungers is caused by the heating of air that is suddenly compressed to 1 atmosphere, and the friction from air moving past at hundreds of miles per hour. However if only a little bit of air is involved, the heating might happen at reasonable rates. Without parameters that I don't have and a fluid dynamics simulation that I am not prepared to do, I can't tell. That said, I can tell you that the rate of heating per volume (and therefore the ease of dealing with it) goes down as you make the plungers larger, up as you speed up, and down as you reduce air pressure. So if you have little enough air and large enough objects moving through it, the heat generation gets more and more reasonable.
And about your PS, I'm trying to guess at how Elon gets around right of ways. Underwater is not part of my basic idea at all.
Musk's pauses indicate to me that he was internally asking himself whether his proposal is somehow like the Jetsons tubes. That doesn't necessarily mean he was looking for literal parallels. And if he were, I don't see why "pneumatic tube that sucks a single person up" would turn into just "tube" to a layman, either. He's also mentioned the Concorde and a railgun. How do those fit into your vision? Will a layperson see those in your car/tunnel system?
I'm not just saying "pointless complexity". I'm telling you why it's pointless. Your system is certainly more complex than an evacuated tube, and without a compelling reason to embrace such complexity, then it is indeed pointless.
Let's step back a bit, though. You're saying that your system is easier to implement than an evacuated tube. Let's assume that's true (though I don't think it is). Is it also easier to implement than a partially evacuated tube? Those are expected to travel at or past Mach I, and don't need the car to move nearly 100% of the air in the tunnel. How is your proposal superior? You talk about using the plungers to evacuate air, but the energy cost to remove the air must still be paid. Your 70 ton slugs might carry more momentum, but a car in a partially evacuated tube wouldn't need to displace all the air it encounters, and so wouldn't need as much momentum. You're constantly paying the cost to push around a bunch of air. That cost is largely avoided with a more traditional design.
I'm not certain about your idea that you don't need a tight seal. Yes, at Mach I a gust would blow pretty hard. But in a near-vacuum you'll not be blowing much air. Remember that you're going to have to blow with >1 atmosphere of pressure to push anything out and past the flaps. Also remember that this "gust" is the result of resistance caused by the giant plug you're pushing down the tube, resistance that another design might just avoid or at least minimize. And any air not evacuated by the "gust" will contribute to further drag as the plunger pushes past it.