We have recently proposed a classification of exact least-squares modeling methods. One of the most promising subset of these algorithms are based on so-called ladder form realizations. They appear in many contexts such as scattering and network theory. In addition, they have several other nice properties and advantages, such as lowest computational complexity, stability "by inspection" properties and relations to physical properties such as reflection or partial correlation coefficients. We shall present new examples of our new exact least-squares recursions for ladder forms, such as Covariance Ladder Form that is equivalent to the so-called Covariance Method or Recursive Least Squares.