This chapter presents an intelligent metasearch engine that can recommend a user’s next hyperlink access and relevant paragraphs extracted from metasearch results. The proposed design is based on the primacy effect of browsing behavior, that users prefer top ranking items in search results. Three search methods were implemented in this engine. First, the search engine vector voting (SVV) method rearranges search results gathered from six well-known search engines according to their weights obtained from user behavior function. The hyperlink prediction (HLP) method then arranges the most likely accessed hyperlinks from the URLs in SVV search results. Finally, the page clipping synthesis (PCS) method extracts relevant paragraphs from the HLP search results. A user study indicated that users are more satisfied with the proposed search methods than with general search engines. Moreover, performance measure results confirmed that the proposed search methods outperform other metasearch and search engines.