That logic opens a metaphysical can of worms. If a mouse = a human, does a roundworm = mouse = human? What is the irreducible essence of human-equivalent life? What are the proofs behind this metaphysical value point, and why should sufficiently large creatures, or creatures that feel pain, or creatures that think, all get those points in a 1:1 ratio? Why not other creatures, or why not a different ratio? To say, one way or the other, "because that is what is right," would be a tautology. Furthermore, under that ethical system, killing an uncontainable fox in a henhouse is morally virtuous, equivalent to saving 10 humans. Or is it? Those 10 hens are causing an insect genocide, by consuming tons of plants that otherwise would have sustained thousands of insects.
Ethical standards are both arbitrary and necessary. They're arbitrary because we choose them, and necessary because they optimize for our goals. Those goals are themselves implicit in our choice of ethical standards. My ethical system assigns the highest weights to human health and happiness. That also takes into account climate change, because I temporally discount large adverse climate events from the future and consider their effect on our progeny. I don't disregard animal life, I just place human life at the top of my ethical pyramid. If you value human and rodent life equally, and therefore human and rodent suffering equally, then I believe your ethical standards are as arbitrary as mine. You just have a different optimization function, and it is at odds, for better or worse, with that of most of humanity.
What we can be most certain of is, generally, the revealed ethical system of most cultures, as manifested by their customs and laws (excluding the myriad effects of autocratic regimes and the cultural stunting that they entail). We can also be sure of our physical location, which determines the cultural mileu and ethical standards which are used to judge our actions. Furthermore, through empirical studies, we can measure the alignment of ethical systems to their goals and adjust course. We cannot, however, find an objective ethical system. It does not exist. We should not be bothered by that any more than we're bothered by the lack of a magic genie lamp or Velveeta volcano. All of that is to say, your equation of: cost of mouse suffering > widespread benefits of resulting drugs, is simply a preference. It can be restated as "I and a few others who share my views prefer that the lab mice not suffer, even if the result helps millions of humans." Even if you actually believe that the mouse testing is only wrong because it is in pursuit of something ostensibly cosmetic, or, as you stated, because we have made life more difficult for ourselves through over-abstraction and animals shouldn't suffer the consequences, your ethical standards are not any less arbitrary.