Accurate software cost estimation is a vital task that affects the firm's software investment decisions before committing required resources to that project or bidding for a contract. This study proposes an improved Case-Based Reasoning (CBR) approach integrated with multi-agent technology to retrieve similar projects from multi-organizational distributed datasets. The study explores the possibility of building a software cost estimation model by collecting software cost data from distributed predefined project cost databases. The model applying CBR method to find similar projects in historical data derived from measured software projects developed by different organizations.