Well a lot of "normal" people do answer yes to all the questions. I would agree with you if the test was so narrow that e.g. only 1% of people responded to it. Then it would not be a social phenomena worth looking into. But a lot more people than a mere 1% agree with the statements, so there is something worth researching there.
So, if you agree the test measures something concrete. Well, then if you look at the guy's book, all his subsequent research has been "well, how much does that predict how people will act or react to a huge bunch of different things". Knowing how people who score highly on the test will act in different situation has value in itself, as soon as you accept that those people are fairly common in society. It also has implications for job selection. High RWA people might be good in particular jobs where they need to adhere to rules but aren't given any authority over other people.
If you just say "yeah, nazi-types are bad, so therefore we don't need a test to detect nazi-type people because we already know they're bad" that basically make no sense. You don't study something by not having means to detect it. How do they tick? They're pretty common, so how can you best work with them, even harness their personality traits in a positive way? You can only ask those questions once you have a metric for finding the people you're after. if you can tell them apart you can do research and make inferences, instead of pure guesswork and just shrugging your shoulders as to why people in the same experiments do different things.
One example of this research which has real applications is that RWA-types like to conform. Altemeyer did an experiment where he opinion surveyed his class, then he presented supposed "average opinions" as fact, and then re-tested the group. The high-RWAs moderated their answers towards the "average" answer twice as much as low-RWA types. So it turns out there's useful information right there about the "Nazi-type follower" if you want to call them that. If an authority figure presents a new "norm" to them, they adjust their behaviour to fit that norm, even if that new norm is far more "liberal" than what they thought before. This suggests many conservatives who say they support gays are doing more than just lip service. If their own party leaders and the TV are saying that accepting gays is the norm, then RWA types are the ones most likely to be influenced by that. So it's much more nuanced than saying they are reactionary nazis. They're more like "right-wing sheeple" that can be controlled once you know what makes them tick.
A really amusing one was when altemeyer told a class about the RWA scale, and got opinions on where they'd like to be on the scale. People who previously tested low to medium on the RWA scale both agreed they'd like to be on the low-end. i.e. they saw not being authoritarian as a good thing. But the high RWA types professed the wish to be medium-RWA. i.e. in the middle of the scale. So they're so conformist that they want to be in the middle of the scale of conformism. Which is highly ironic. But you can see how someone who's obsessed with conformism and not standing out can be easily manipulated.