Our L1 defense is actually incredibly good. A human will undergo about 10^16 cell divisions over the course of their lifetime. Around 10^3 to 10^6 of those divisions will result in a mutation that gets past the L1 defenses and need to be eliminated by the T cells. It's not generally easy to make dramatic improvements to something with a 99.9999999% success rate.

The immune system is pretty good too, which means any given improvement to the replication system is, all else being equal, probably going to prevent mutations the T cells would already handle. If you need to do the research to figure out what's getting past the immune system anyways, and improving the immune system is lower hanging fruit, it's the logical place to start.