Cristobal, the question posed by your article is more profound than many will realize. Given that selection pressures have not exactly favored the development of reason over the course of our history as a species, it's astonishing that some tiny percentage of the population should be capable (at least occasionally) of demonstrating such a trait.

As for the inverse correlation between IQ and religiosity, it's important to distinguish between limited cleverness and generalized cleverness. By way of analogy, a fighter pilot is highly skilled in one domain, but may have negligible competence with regard to playing a musical instrument, speaking foreign languages, or being able to make correct inferences when presented with metabolic pathway information. So it's not really surprising that occasionally a very rare religionist is able to demonstrate some cognitive skill associated with IQ. What counts is their in ability to utilize this limited domain-specific talent by extending it to a wider problem domain. And that is strikingly absent.

Religion requires a fundamental inability to reason, supported by an equal inability to perform even basic consistency-checking (is belief A consistent with belief B?). The lack of necessary neural pathways is what ensures that religious people are always unable to follow a logical argument to its inevitable conclusion but must instead forever swerve off into irrationality and obfuscation. This makes any attempt to reason with a religious person completely pointless.

As the saying goes, "if you could reason with religious people, there wouldn't be any religious people."