Salazar's reports were argued over inside the Inquisition, against colleagues who did not accept them, and they largely won. In 1614 the Suprema issued revised instructions for witchcraft cases which reflected his findings: confessions were not to be taken at face value, accusations by children were to be treated with great care, cases were to be reviewed centrally, and the tribunals were in effect discouraged from proceeding. The Spanish Inquisition, whatever else it did afterwards, is not a significant executor of witches from that point on, and Spain and Portugal have strikingly low totals for the period when much of northern Europe was at its worst.
Two things follow, and both cut against the popular picture. First, a strong central reviewing authority reduced witch executions where it existed, because it took the decision away from local courts responding to local panic. The regions with the heaviest prosecution are generally the ones where small jurisdictions could try and burn without review. The Inquisition's reputation makes this uncomfortable, and it is nonetheless what the records show.
Second, argument from inside worked, at least once. Salazar was not a sceptic in the modern sense and did not deny that the Devil acts. He simply insisted on evidence, and on testing what he was told against what he could find. Gustav Henningsen's study of the episode gave him his English name, the witches' advocate, and reconstructed the visitation in detail from the tribunal's own papers.