The immune system is amazing. You only have about 30,000 genes, but your immune system generates tens of millions of kinds of antibodies. Clearly, some sort of cut-and-paste, shuffle-and-deal technique has to exist in there.
In the 1960's, science got to the point where we could study the molecular-level details of such things. It turned out to be extremely complicated, and some creationists use that complexity as an argument against evolution.
Scientific efforts have recently paid off. There is now a detailed theory as to how the system could have been started by a transposon. And, there is now experimental evidence that elements of the immune system have many suspicious resemblances to exactly such a transposon [1]. The technical details are rather difficult going, but see [2] for a relatively accessible overview. Notice that this complexity is only to be expected. After all, Behe talked about the immune system specifically because it was complicated.
The biggest single coincidence is that the genes which generate the RAG1 and RAG2 proteins turn out to be collectively a transposon. That is, they (and their locations) are exactly what would be expected if they used to be a transposon that landed inside a gene controlling the surface of white blood cells (to be exact, a membrane-spanning-receptor gene used by lymphocytes.) From other evidence, this event happened about 450 million years ago. The transposon itself may have been due to mutation, but there are strong suspicions that it jumped from the DNA of an infection (a virus, germ, or even fungus) across into the DNA of the host.
The report of the breakthrough says "some aspects of this model ... require experimental confirmation" [1]. This means that the researcher is making predictions and is therefore doing science. Further results are expected.
[1] Transposition mediated by RAG1 and RAG2 and its implications for the evolution of the immune system Agrawal, Eastman, Schatz Nature vol. 394 pp. 744-751 (20 August 1998). See also review on pp. 718-719 by Ronald Plasterk.
[2] The Accidental Immune System, John Travis, Science News Vol. 154 p.302-303 (7 November 1998)