The Artificial Gorilla Groups Optimizer (GTO) is a novel metaheuristic algorithm that takes its cues from the collective intelligence of wild gorilla troops. Although it has shown promise in solving various real-world applications, it may still get stuck at local optima and premature convergence when dealing with more complex optimization tasks. Because of these shortcomings, this research proposes a new modified Gorilla Groups Optimizer (mGTO) method, employing a set of operators to achieve a more stable balance between exploitation and exploration. These operators are the elite Opposition Based-Learning (EOBL), Cauchy Inverse Cumulative (CICD) Distribution Operator, and tangent Flight (TFO). Each operator is used to achieve a specific task during the search process. The EOBL aims to enhance the diversity of the population, which leads to discovering the feasible region. Whereas the integration between CICD and TFO is used to improve the population’s exploitation ability, which leads to an increase in the convergence rate. To validate the efficiency of the presented method, called mGTO, a set of experimental series is conducted using the CEC2020 benchmark and constraint design engineering problems. In addition, its applicability is assessed by implementing mGTO as a feature selection technique and applying it to improve the classification accuracy of sixteen datasets. The results of mGTO are also compared with those produced by other well-known meta-heuristic techniques. The statistical validity of the performance is also verified using Wilcoxon’s rank-sum test. The experimental results and comparison analysis reveal the consistent and better performance of the proposed mGTO method to solve optimization problems.