VSEPRT is problematic and some how get stuck as you go in to more detail analysis. One of the classic examples would be the inequivalent of the lone pair on oxygen of water molecule. Take a look here if your are interested :D
But I still think VSEPRT would be a good start for learning chemistry at the very beginning
Actually with the current technology, we can really visualize molecule by atomic force microscopy(AFM)
Here is an example of pentacene nicely demomostrated by IBM with AFM
Frankly, I dont know much about this technology so may be see if some expert can explain a bit lol.
Did you read the article you linked? The article was against retiring the VSEPR model:
"Critical analysis, which includes a detailed examination of the photoelectron spectrum of methane, reveals these premises (premises on which the anti VSEPR advocates established) to be ill founded and inconsistent with modern electronic structure analyses. Placed in a modern context, the hybrid orbital concept helps to familiarize students with the methods of working chemists, foster construction of a deeper, more interconnected understanding of chemistry and its connection to the laws of nature, and provides a secure foundation for more advanced chemistry classes."
I do agree that VSEPR is a good model to be used for into chemistry classes. There are even exceptions to some molecule structures within VSEPR theory itself, for example The gas phase structures of the triatomic halides, AX2E2 molecules which are linear rather than bent, and AX6E1 molecules, which are octahedra rather than pentagonal pyramids. So the theory does recognize that there are exceptions to the rules. I wouldn't say that just because in advanced and specific cases that the theory should be done away with!
My university has an AFM lab! But it is only open to graduate students and really only works with very specific cases.
Oh! I am terribly sorry that I mistakenly attached the old publication by F. Weinhold in 2012, as he made a few comments on VSEPRT.
The one I was originally trying to show you was this one which is done by Weinhold in 2014.
"VSEPR-style representations of orbital shape and size are shown to be fundamentally inconsistent with numerous lines of experimental and theoretical evidence, including quantum mechanical ‘‘symmetry’’ principles that are sometimes invoked in their defense. VSEPR-style conceptions thereby detract from more accurate introductory-level teaching of orbital hybridization and bonding principles, while also requiring wasteful ‘‘unlearning’’ as the student progresses to higher levels"
This publication was made in to a lecture material in the Harvard University which I originally came across in the online teaching materials