For example, make a flimsy structure out of thin steel foil, then pressurize it to give it far more strength than it would otherwise have.
Same as the strength of a coke can.
Except with steel as the tensile material, you can probably use pressures up around 1000 psi, making the whole structure have an amazing strength to weight ratio.
Congratulations, you have created a bomb.
You also need pressure regulator valves in case of fire.
And you might want to make the whole thing double or triple skinned so the highest pressures can be deeper inside where they are less likely to get punctured/less vulnerable to corrosion damage/crash damage.
Just for some reference, 1000psi is the kind of pressure you get ~700m depth (2300ft), double what a submarine would regularly withstand. Pressure in your average propane gas tank probably peaks around 200psi in high temperature. That steel isn't flimsy, it tends to get bulky and heavy even with creative ribbing and reinforcement.
It has to make a lot of sense practically and economically to consider such an option. So it ends up being used in places where weight or cost matter less.
This highlights that high pressure and "thin steel foil" do not go well together and don't scale in your favor. Pressure makes any failure considerably more dangerous.
Unmanned aviation comes to mind, but the ratio between structural mass and battery+payload is already so low, even a zero-mass structure would not make a meaningful performance difference. It's all about the battery tech.
Scaling up foil boating perhaps? Carbon fiber construction would certainly appear like a good candidate for taking advantage of being pressurized, and I perhaps naively take it as a given that materials is the major limiting factor in scaling up.
You might not design for that case. A suspension bridge can't survive a tower falling over, or a cable snapping.
The pressurized structures wouldn't be able to survive leaks over X size, and for leaks under X size, compressors would be used to maintain pressure.
"The tank of the Atlas consisted of stainless steel which was no thicker than a dime at any point. Generals, and later congressmen, worrying about what would happen if somebody dropped a wrench on such a tank were conducted to a test version of the "stainless steel balloon" that was stiffened by pressure and offered a choice of assorted mallets to see whether they could dent it. In each case the general or congressman grew tired before he had even succeeded in producing a mark that could be seen."
Willy Ley's "Rocket's, Missiles and Men in Space"
Fueled rockets are treated like live bombs. No one around except for people who really need to be (like astronauts).
I'm assuming you're referring to the unnominal situations resulting in the unscheduled rapid disassembly procedures? I wonder how many people watch launches with similar intent as NASCAR races where they're just looking for the crashes or hockey games looking for the fights?
But for, say the JWST launch, it really wasn't with the same intent.
In the former, it is just SpaceX way of doing business, no big deal, no significant payload being destroyed and the next rocket is on the line. The latter is the result of billions of dollars, many years of work and great scientific promises that could be blown to pieces.
And of course, for manned flight, who in their right mind wants astronauts dead?
Which begs the question, who wants to fly Starliner?
I think this is the failure you're referring to: https://www.youtube.com/watch?v=imkdz63agHY
And here's a picture from ULA/Tony Bruno showing a Centaur tank in their factory, being supported by a rig: https://twitter.com/torybruno/status/1435553806792728579
For reference, a steel or aluminum SCUBA tank is usually filled to 3000psi and when I fill steel tanks I can hear and sometimes see the metal start to slightly balloon. This is normal because steel is not brittle but it is very strong.
There is a lot of rules put in place for safety though. A tank has to be visually checked by a professional every year and hydro statically tested every 5. The danger is any rust/corrosion weakening the metal and causing it to rupture. More often seen in aluminum tanks, which is one reason I prefer steel.
We use it because it's cheap and well tested. But we are probably better of with a layered combination of a few different materials. Steel may not even enter it.
Layers of dissimilar materials are often problematic due to galvanic corrosion, different rates of heat expansion, and water intrusion between layers.