The evidence certainly points in that direction, but it's very far from conclusive. Going back to the article linked on the last page, it appears that there have only been three Neanderthal specimens from which mDNA has been extracted. These three exemplars were then compared to a vastly broader sample of modern mDNA, and conclusions were drawn from that.
You have to be skeptical of an study that bases its conclusion on such a narrow sample of Neanderthal mDNA because you cannot exclude the possibility that there are other Neanderthal mDNA sequences that existed at the time of interbreeding, but are not present in the three Neanderthal samples to which modern mDNA is being compared. For example, note that the article compared the Neanderthal sample from the Neander Valley to the one taken from the Caucasus, one had ~27 differences from the modern sequence, while the other had only ~22, or about 25% less. Therefroe, a greater sample of Neanderthal mDNA could reasonable be expected to have many other mDNA sequences not found in the three samples tested, and some of those might have been detected in modern mDNA. Based on the data the article presents, you cannot exclude that possibility.
I continue to question the validity of any comparisons to modern H. sap DNA. The comparisons should be to similarly old H. sap. DNA.
Plus the genetic differences appear not to be great enough to guarantee inter-group infertility, even at this chronological remove.