If you were to do this by hand, would you search every book and paper ever written? That is not feasible so you have to make a trade-off.
For alternatives to vector search, see GraphRAG and AST parsing; e.g., https://vxrl.medium.com/enhancing-llm-code-generation-with-r... or https://github.com/sankalp1999/code_qa
That's what google scholar is for. Use it to find the meta analysis papers and go from there.
Which incidentally shows why RAG just means vector store + embedding model, since your definition means different things to different people and an implementation can't exist until we figure out AGI.