Since the .children uses \ "_" which skips Text nodes, there is no easy way to gather just text nodes or more generally, iterate over all nodes, including Text nodes, using the combinators. Since the signature of XmlReader.read pushes a NodeSeq argument, you are a bit stuck.
scala.xml Node.descendant returns all children (entire tree) including Text nodes, but Node \ "_" does not. So you are forced to hop through nodeReader. This is not urgent, but it seems like a better .children that is comprehensive would be nice.