The presence of low quantities of water vapour can seriously affect the kinetics of reduction of iron oxides when they are used as catalyst or to store and/or purify hydrogen from streams in the steam-iron process. Only 5% (v) of steam should be enough to inhibit the complete reduction of the solids. Since steam is a product of the reduction reaction, small amounts of water present in the reactive atmosphere can slow down the reduction itself. To account for the effect of the steam pressure during the reduction stage of the steam-iron process, two approaches have been considered and the resulting models, i.e. 'competitive model' and 'inhibitive model' have been tested against experimental measurements. Both models are based on the known Johnson-Mehl-Avrami-Kolmogorou (JMAK) theory. The 'competitive model', accounts for the discretization of groups of moles of iron oxide/iron reducing and oxidizing with their own reaction rates. By using the kinetic parameters obtained from independent reduction and oxidation processes, this model is not capable of predicting properly the behaviour of the solid subjected to successive reductive and oxidative cycles. On the contrary, the 'inhibitive model', which takes into account the hydrogen and water vapour partial pressures in a Langmuir-Hinshelwood type kinetic constant dependency, seems to be very appropriate to predict correctly the effect of the presence of water in the reducing atmosphere. Copyright (C) 2011, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.