Real-life optimization problems require an effective mechanism that completely utilizes the search space to obtain optimal solutions. A designer has always an opportunity to propose a new effective solution technique to address this issue. In this order, this paper presents a new upgraded redefined model of nature-inspired state of art meta-heuristic firefly algorithm (FA). FA is centered on swarm intelligence, which is motivated by the flashing pattern and behavior of fireflies. However, for a few instances, FA has a tendency to trap in local optima and it exhibits slow convergence. The proposed model is enabled through Time-varying Inertia Weight (TIW), Opposition Based Learning (OBL) and hybridized with sine cosine operators to get updated positions of search agents. A set of twenty-two classical benchmark function problems with numerous range and features are employed to prove the efficacy of the proposed model. Also, the effects of the above-mentioned modifications during the whole optimization routine are analyzed through different statistical and numerical analyses. The result analysis proves that the proposed modifications make FA more compatible. The proposed model is also tested on the strategic bidding problem of the power market of two different test systems with single and multi-trading hours. All the reported results confirm the supremacy of the proposed redefined model of FA.