Mostly good points in your article, but this one (these three) isn't a good example of "because we've always done it that way".

We kill billions of animals for food every year because we're omnivores and have both a dietary need for and a taste for the kind of nutrients easily obtained from animals. And we are highly cruel to many of them because it improves efficiency and we haven't figured out how to get a remotely stable large-scale civilization that doesn't prize efficiency. If we stopped doing it, and forgot we ever did, we'd do it all over again without some other explicit reason not to. It's almost completely driven by need and desire, not by custom. Maybe we can stop it by custom, if we collectively decide to.

We kill relatively few animals for clothes. Humans need clothes in most environments (and in those where the don't, really, they still tend to prefer them for modesty, artistry, and sometimes function). But these days we've found superior or at least adequate alternatives to most animal-based clothes, and/or use surplus material from animals we were going to eat anyway. So this is mostly a non-issue. Most of the remaining cases are again where it's driven by need, not custom. (Leather would be an exception if we didn't already kill so many animals for food.)

The animals killed for research are done almost entirely because there is no other adequate way to learn about various topics that we deem important. Although in the past experiments were done with a great deal of haphazard indifference, these days there are careful reviews of protocols, and also, most everyone recognizes that animal experiments are expensive, and they would get their results faster and less expensively if they didn't need animals. But, alas, if we're going to understand the immune system (so we don't die in pandemics), regulation of blood metabolites (so we don't die of diabetes), etc. etc., we are absolutely required to do some experiments on animals. Experiments on cells or yeast or something is not conclusive (though whenever possible hypotheses are initially validated in systems like these where you can do fast, inexpensive experiments), and experiments on humans are too limited and too slow. Again, this is driven not by custom but by need, and unlike with food, there is no (morally acceptable) alternative save stay ignorant and die from diseases that would be preventable if we understood what was going on.

Some things are just matters of tradition. But these are not.

Note also that tradition is slightly better than circular logic in the case where not all things work. If the tradition is a tradition to do something that works--maybe poorly, maybe with drawbacks, but it does accomplish some goal thought to be important--then just randomly doing different things is not an equally good choice in many areas. Doing things differently may be vastly more likely to produce far worse drawbacks and no benefit than an overall improvement.

So when examining tradition, we must remeber that we don't have to keep doing things a certain way, but also that maybe there was a good reason they were done this way. Simply changing them as we see fit, without fully understanding why the tradition was established and what it accomplishes and what it doesn't, is not necessarily wise. This isn't to say that we shouldn't change traditional things--to the contrary, we should, when appropriate; we should just be sure to do the work of understanding in full what the tradition was accomplishing.