A second look at the second messenger hypothesis