Fiber-reinforced composite structures are used in different applications due to their excellent strength to weight ratio. Due to cost and tool handling issues in conventional manufacturing processes, like resin transfer molding (RTM) and autoclave, vacuum-assisted resin transfer molding (VARTM) is the best choice among industries. VARTM is highly productive and cheap. However, the VARTM process produces complex, lightweight, and bulky structures, suitable for mass and cost-effective production, but the presence of voids and fiber misalignment in the final processed composite influences its strength. Voids are the primary defects, and they cannot be eliminated completely, so a design without considering void defects will entail unreliability. Many conventional failure theories were used for composite design but did not consider the effect of voids defects, thus creating misleading failure characteristics. Due to voids, stress and strain uncertainty affects failure mechanisms, such as microcrack, delamination, and fracture. That’s why a proper selection and understanding of failure theories is necessary. This review discusses previous conventional failure theories followed by work considering the void’s effect. Based on the review, a few prominent theories were suggested to estimate composite strength in the void scenario because they consider the effect of the voids through crack density, crack, or void modeling. These suggested theories were based on damage mechanics (discrete damage mechanics), fracture mechanics (virtual crack closure technique), and micromechanics (representative volume element). The suggested theories are well-established in finite element modeling (FEM), representing an effective time and money-saving tool in design strategy, with better early estimation to enhance current design practices' effectiveness for composites. This paper gives an insight into choosing the failure theories for composites in the presence of voids, which are present in higher percentages in mass production and less-costly processes (VARTM).