Honda's not in dire straits YET. The three flops mentioned in the original article are all niche vehicles, not their meat-and-potatoes vehicles (Fit, Civic, Accord, CR-V, Pilot, Odyssey). HOWEVER, I am concerned looking at recent Honda offerings that Honda is starting to lose its focus. A company's lineup can't be all things to all people. You have to know what your strengths are and what your weaknesses are.
If I was running Honda, I would run every new design through the following litmus test. Cars that passed would have, I think, a distinct "Honda-esque" identity that helped build the brand into what it is today.
Not everyone is going to like the results of this litmus test. That's fine. Honda doesn't have to be the best-selling car maker on the planet so long as it makes a profit on each car it sells. Better to have a coherent vision and know your base than to dilute your product by being too many different things to too many different people. I think the best examples of recent cars that fit this litmus test are the current Fit, the current Civic, the recently-deceased S2000, the first-gen TSX, and last but certainly not least, the NSX (the perfect halo car for Honda's traditional virtues). Not right for everyone, but nevertheless, excellent and successful cars.
Under my litmus test, the car MUST have the following strengths
Relatively light for its class
Relatively agile for its class
Relatively fuel-efficient for its class
Relatively high-revving for its class
Simple, modern, aerodynamic exterior design
Function-over-form interior design
Over time, good mechanical and electronics reliability
Under my litmus test, the car MAY have the following weaknesses and still pass the test
Relatively high road noise for its class
Relatively low torque for its class
No traditional good looks (inside or out) if traditional good looks don't make as much sense from an engineering/pragmatic standpoint
Over time, squeaks and rattles develop