Lung diseases such as Pneumonia, COVID-19, and Tuberculosis (TB) are major contributors to global mortality, making early and accurate diagnosis crucial for effective treatment. This paper introduces a novel Hybrid Nature-Inspired Optimizer – Convolutional Neural Network (HNI-CNN) model for the classification of lung diseases from chest X-ray images. The model integrates Particle Swarm Optimization (PSO) and Firefly Algorithm (FFA) to form a hybrid feature selection method, which optimizes the extraction of relevant features, significantly enhancing the classification accuracy. Additionally, Principal Component Analysis (PCA) is employed to reduce dimensionality, ensuring the model focuses on the most informative features. Tested on the Kaggle lung disease dataset, the HNI-CNN model achieves state-of-the-art performance, with an accuracy of 98.0%, precision of 96.8%, and recall of 99.0%, outperforming leading models such as NasNetMobile, MobileNetV2, and EfficientNetB1. This research presents a significant advancement in intelligent diagnostic systems by offering a more reliable and efficient method for feature selection and classification. The hybrid optimization approach not only improves performance but also enhances the robustness of automated medical diagnostics, making it a valuable tool for expert systems in healthcare.