On pg. 33 and in fn. 16 thereto I cautiously mention the Malayo-Polynesian languages Kelabit and Madurese as potential parallels (currently in the case of Kelabit, or formerly in the case of Madurese) to the three-way laryngeal process contrast (voiceless, voiced, breathy) reconstructed for PIE. Since writing that, I have looked into these languages a bit more and am more confident that they really are parallels to the PIE system. My discussion can be found in the slides posted as The Cao Bang Theory (scroll down). But as Eduard Fraenkel said somewhere, "the world is full of pseudo-parallels."