After much research, I've concluded that there are only 2 jackets. The vintage one Kaplan found in the boutique to base the design off of (sold to Tom via screenused.com), and the actual screen used jacket as shown below. (Fabricated by JAL for costume designer Michael Kaplan)
I guarantee you, these are all the same jacket. (photos are in chronological order. Notice how different the red color appears in each pic)
In every instance the jacket appears in the movie, the top broken button is exposed and the lapel is folded above it. The jacket also appears this way for the Bee Healthy shoot. However in the promotional photo-shoot with Matt Welch (which took place during production inside the paperstreet house), you can clearly see the jacket's lapel was refolded and creased lower, over the broken button. (Maybe Matt just didn't like the look of the broken button and wanted to show off more of the moto shirt.) You can see the higher crease lines in the lapel in these photos. Here is one for example:
It's around this time that I suspect the stitching started to come undone on the pocket flap. Then in the photo-shoot with Steven Klein (which took place after production) the lapel was folded back to it's original higher position above the broken button as seen on the cover of W magazine. If you look closely here you can see the lower crease marks on the lapel:
A few years later in Aug of 2004, this photo was taken of the full costume to promote the "50 designers/50 costumes" exhibit:
But before it went on display for the exhibit itself, the lapel was folded once again higher up, possibly by Kaplan himself making it look as he remembered on set as you can see him making adjustments to the jacket in this photo taken in Sep of 2004:
However, In all the following instances where the jacket has been on display since, the lapel has once again been creased lower and has remained that way for the last 10 years.
Clear example of the double crease:
Case Solved! It's all the same jacket.
Anyone still not convinced? I have further evidence . . .