Cross-Project Defect Prediction (CPDP) has emerged as a hot topic in software engineering since it fosters software quality across multiple projects. However, it is still a challenge owing to divergent data distributions and differences in features between projects. This paper presents an innovative approach that implements Fuzzy Neural Network Logic to enhance prediction performance through the meaningful handling of uncertainty and imprecise data. Evaluation on benchmark defect datasets shows an increase in accuracy and generalization compared to conventional machine learning techniques. Major improvements include better defect classification, adaptation to cross-project scenarios, and high confidence in prediction results.