This paper proposes a cascading outages model which captures power system operations, protections, automatic regulations, and remedial action schemes. It is based on linear implicit AC PF, which provides a tradeoff between computationally expensive AC PF and approximate DC PF. Combinations of several types of failures and interactions in blackouts are represented, i.e., cascading line overloads, large frequency deviation, reactive power problems and voltage drop. Statistical comparisons demonstrate that the linear AC PF-based model enhances the accuracy of the power flow solutions as compared to the DC PF with minimum increase of computation time. The case study illustrates that the AC PF-based model provides a more realistic estimate of the consequences of cascading events as compared to the optimistic estimates by DC PF-based models. The computational efficiency and solution accuracy of linear AC-PF model can be exploited in the estimation of the cascade risk for a large number of scenarios.