Software may prone to different bugs at different stages of software development life cycle. There are various ways to find and rectify these bugs. Testing is one of these techniques to find out bugs in the software. These bugs and problems can be efficiently found with the help of software testing. Testing can be done from different perspectives; the software can be tested for its functionality, for its performance, or for some other purpose. Testing can be done manually or automated. To save time and cost it is needed that software testing might be performed with the help of computer. Test coverage and accuracy of results is also increased with the help of automated testing. Many automated testing models have already been proposed in this regard. In this research, two models have been proposed for automated testing of software. For automated testing of nested if-else structure, binary tree has been used. Second model has been used for automated testing of ?for? loop statement logic. Two tools in java have been developed to implement and validate this research. After comparison with existing techniques and models it is found that proposed models are better than previous models in terms of code size, code complexity, complexity of applied technique, and algorithm completeness. Other models in previous researches have at lest O(n2) runtime complexity whereas proposed models have optimized runtime complexity.