I would relabel PTSD to "intense aversive conditioning", just to avoid loading the notion of PTSD with all the professional psycho-babble. Then I would note the frequent observation that the stimulus with the aversive conditioning "moves" from short-term memory in the Limbic System to long-term memory in more as to a "holographic" model of the brain, which includes structures outside the Limbic System.
The extinction of the effectiveness of the stimulus in intense aversive conditioning does not occur in humans. Attempts to reduce the effectiveness of the stimulus often does more harm than good, such as treatments that destroy the associated verbal memories (and too much else of unrelated elements), but necessarily leaving many elements of the "holograph" that leaves a non-verbal "mystical" set of elements often then labeled as another "mental illness".
Experiments to stop the short-term memory from becoming a long-term memory (whether by drugs, to electro-convulsive treatments, etc.) have not been very reliable or effective, and are probably immoral and/or unethical in use on humans.
In my experiences with intentional strong aversive conditioning, later exposure to the stimulus elicits the conditioned effect strong enough to additionally reinforce the conditioning over any degree of accompanying extinction. The stimulus can "morph" into different "modalities" (such as, even sufficiently "thinking" of/about the original stimulus used will start to elicit the effect).
Tadzio