The increase in tight reservoir exploitation projects causes producing many papers each year on new, modern, and modified methods and techniques on estimating characteristics of these reservoirs. The most ambiguous of all basic reservoir property estimations deals with permeability. One of the logging methods that is advertised to predict permeability but is always met by skepticism is Nuclear Magnetic Resonance (NMR). The ability of NMR to differentiate between bound and movable fluids and providing porosity increased the capability of NMR as a permeability prediction technique. This leads to a multitude of publications and the motivation of a review paper on this subject by Babadagli et al. (2002). The first part of this presentation is dedicated to an extensive review of the existing correlation models for NMR based estimates of tight reservoir permeability to update this topic. On the second part, the collected literature information is used to analyze new experimental data. The data are collected from tight reservoirs from Canada, the Middle East, and China. A case study is created to apply NMR measurement in the prediction of reservoir characterization parameters such as porosity, permeability, cut-offs, irreducible saturations etc. Moreover, permeability correlations are utilized to predict permeability. NMR experiments were conducted on water saturated cores. NMR T2 relaxation times were measured. NMR porosity, the geometric mean relaxation time (T2gm), Irreducible Bulk Volume (BVI), and Movable Bulk Volume (BVM) were calculated. The correlation coefficients were computed based on multiple regression analysis. Results are cross plots of NMR permeability versus the independently measured Klinkenberg corrected permeability. More complicated equations are discussed. Error analysis of models is presented and compared. This presentation is beneficial in understanding existing tight reservoir permeability models. The results can be used as a guide for choosing the best permeability estimation model for tight reservoirs data.