The objective of the present paper is to incorporate the effect of plastic anisotropy on the fracture modeling of aluminum alloy 6061-T 6 sheets. Six different types of tests were performed to fracture, including tensile tests on classical dog-bone specimens, flat specimens with cutouts, plane strain grooved specimens, and punch indentation tests on circular blanks. A limited number of shear/tension tests on butterfly specimens were performed on a dual-actuator loading frame. Plastic properties were determined from the dog-bone tensile tests, and were verified by the remaining tests. It was found that the sheets exhibited little planar anisotropy but substantial out-of-plane anisotropy, characterized by the Lankford parameter, r. A comprehensive numerical analysis of the experiments revealed that the Hill 1948 quadratic anisotropic yield model is able to describe, with good accuracy, the plastic response of all five types of tests. Fracture surface strains were measured using a digital image correlation system. Average fracture strains were determined by measuring post-fracture thickness reduction. Local fracture strains were determined by means of an inverse engineering method involving matching the displacement to fracture from numerical simulations to those measured. Possible discrepancies between the magnitudes of the fracture strain in the three above methods are discussed. Based on the previous experience of the investigating team, the experimental fracture data were analyzed within the realm of the three-parameter Modified Mohr–Coulomb fracture model. Using the plane stress anisotropic plasticity equations, the calibrated fracture model was then transformed to the space of the equivalent strain to fracture and stress triaxiality. An alternative representation of the fracture locus in the space of principal strains was also constructed. Other important factors influencing the form of the fracture locus, such as mesh-size effect and solid versus shell representation, were also investigated.