As soon as we start to rely on an idea or toolkit simply because it has always worked we risk falling over. Advertising has always evolved, but now it is doing so quicker than ever, and just because something has worked for 20 years does not mean it is right for this brief, this client or this problem. If you can look at the problem and data that are relevant and then say 'this model definitely still applies' then brilliant, but fail to check and its the first sign of complacency.
There is nothing wrong with using models per se, as long as they help instead of hindering; don't steal time from the real issues; and are always sense checked for their relevance to a brief.
Some agencies have great models that are really just philosophies to drive themselves. It's why I will always prefer BBH's 'When the World Zigs, Zag.' to TBWA's 'Disruption'. The former (at least from my understanding) is a creative philosophy that doesn't get in the way of finding what is relevant; while the latter feels like a process that every bit of work is dragged through regardless of if it needs. [Please correct me if I am wrong.] Not to say it doesn't produce great work sometimes, but it seems like another hurdle rather than an enabler.
Every case is different, what works for one may not work for another.
Not only that, but this issue is also part of the problem involved in the 'advertising is dead/new model agency' debates that keep coming up. If it is dead, the reliance on archive models and ideas is probably a big reason why. The new models need to make sure their ideas are based on what is right, not what is currently wrong. Even the people claiming to look forward seem to be using the past as their start point.
This is why in my mind much of this debate can be summed up by saying: Advertising is not dead, it's just learning where it needs to go. The tried and trusted models aren't always right anymore.
I believe this industry works best (both in terms of results and in terms of creativity) when we treat every problem with an open mind based on the relevant problems/details/insights around it. Not when we stick a proprietary creative/planning system on top of the problem and end up having to solve two different issues to justify ourselves.