Bio-inspired flapping wings have garnered a large amount of attention in achieving the goal of developing energy efficient highly maneuverable hover-capable Micro Air Vehicles (MAVs). Flapping wings are typically very flexible undergoing large deformations, leading to complex nonlinear interactions between aerodynamics and structure. Computational studies are critical to understanding this highly coupled non-linear Fluid Structure Interaction (FSI) problem and to explore the wide design parameter space for developing efficient and robust flapping wing MAVs. Arbitrary Lagrangian-Eulerian (ALE) methods for Computational Fluid Dynamics (CFD) are unfeasible to handle large structural motions and deformations that are seen in flapping wings. Eulerian Embedded Boundary Methods (EBMs) are particularly attractive in this situation. However, EBMs become expensive in viscous FSI problems, since they do not track the boundary layers around bodies because of their Eulerian nature. Recently, the current authors developed an ALE-embedded framework, in which the underlying non body-fitted mesh tracks structure and therefore providing significant computational savings. The focus of the current work is to apply this framework to study highly flexible flapping wings; the initial step being to carefully validate the methodology with an available experimental data. Three different wings of varying flexibility are simulated for a range of flapping frequencies. The calculations show good qualitative agreement with the experimental measurements in predicting the structural deformations as well as the generated thrust. The quantitative differences are primarily attributed to the uncertainties in the structural model. The results indicate that wing flexibility is beneficial for flapping wing aerodynamics, however excessive flexibility can be counter-productive. Flow visualization show the presence of stable leading edge vortex on moderately flexible wings; whereas no stable vortex is evident on highly flexible wing.