Russell Duhon's point here is really interesting and a good call out on the cited research if correct. I have this book in our collection (it was once upon a time a bit of a bible for our industry). I will be taking a closer look too.
I think there's also a difference between the blog's description of situation and Russell's point is that the book cited the improvement when redundant text was matched with verbatim audio and interspersed with animation - while the blog references a graphic on screen alongside redundant text and accompanying audio i.e. all at the same time.
I am interested if Russell's point about the research cited has influenced your thinking Mark? It's always such a discussed area in learning design.