The fiber reinforced polymer composites (FRPC’s) are essential for engineering applications that require materials with low weight, high strength and high fracture toughness. Although FRPC’s are quasi-brittle materials, the failure analysis of large structures made of these materials such as fuselages and wings of airplanes seems to be easily achievable using Linear Elastic Fracture Mechanics (LEFM) given the small size of the fracture process zone in these materials. However, a number of different modes of failure have been discovered in these materials some of which cannot be predicted by LEFM. Moreover, a quasi-brittle model which can predict all of these failure modes seems to be still out of reach. The Cylindrical Microplane Model for FRPC laminae (the model CMM) has recently been shown to fit a wide range of experimental data including uniaxial tension, uniaxial compression, biaxial tension, biaxial compression, biaxial tension-compression and 3 point bending size effect test data with post-peak [1]. However, in real life applications a very large mesh is needed so that each lamina can be discretized individually. For example, in [2], a simple cohesive model is used to accurately (within 9%) capture the strength reduction with increasing laminate thickness in the open hole specimens using 5 million elements. The largest specimen had gauge dimensions of only 50.8×12.7×0.8cm having a hole diameter of 25.4mm. The run time was as long as 60h despite employing 72 CPUs [2]. Clearly a computationally much less demanding but still accurate model at general 3D states of stress is desirable. To this end, in the computationally much more efficient model CMML, the model CMM is lumped at a material point of a shell element for each lamina and their individual responses are enriched by taking into account the interactions between the laminae including delamination at general triaxial stress states. Finally, these stresses are combined together using principle of virtual work to yield the Cauchy stress.