Terminology: type 1 flight, no security. Type 2 flight, security.
A. When bad things happen on type 1 flights, people will say 'thats what they get', and not worry too much. Type 2 flights being targeted have more impact.
B. Type 1 flights will presumably be under more scrutiny as a flight -- they will be 1. more restricted in airspace, and 2. be more likely to be shot down in response to deviant behavior, as such less likely usable to reach bad guy goals.
C. Airplanes are expensive enough that the airlines would need to figure out ways to minimize bad guy impact on thier type 1 flights, rather than just hope the security measures are good enough.
Do people want to be strip searched? Well, let's find out. You can take flight A where nobody is strip searched, or flight B where everyone is (all other existing security measures unchanged).
If nobody (but me) is willing to take flight A, we have the answer we're looking for.
My point was that everyone assumes that nobody will take flight A, but in fact we don't really know.
1. If some flights have security procedures and some don't, obviously all malefactors will be attacking the ones that don't. So they'll be more dangerous than if no flights had security procedures.
2. No one is really suggesting having no security procedures at all, so far as I know. The question is how much security overhead we're prepared to put up with, and how much benefit in safety we're actually getting.
I might be reluctant to get on flight A. But if the great majority of flights had simpler quicker security procedures only -- say, 1/4 of the way from flight A to flight B -- then I bet I'd be about as safe on them as I am now with all flights being like flight B.
Back of envelope calculation. Suppose I value my life at $10M, and my time at $50/hour. Then I should be prepared to put up with an hour's timewasting if it saves a 1/200,000 chance of death. The actual rate of violent death for aeroplane passengers over the last half-century or so has never been higher than about 1/5,000,000 per flight (in the 1940s and 1960s; it's been much lower since then even counting 2001-09-11; even if you include non-passengers, who of course made up most of the 9-11 deaths, the figure is still lower for the 2000s than for the 1940s and 1960s), meaning that I should be willing to put up with about 2 minutes of security overhead per flight. In practice, reducing the amount of security checking would probably increase the danger somewhat, so the correct figure is probably more than 2 minutes. I doubt it's more than 15.