Just pondering this a little. Much like "calibers" for stability factor, this seems to be a rule of thumb that's based on an assumption about an "average" rocket, but in its conception switches from the relevant variable to another one that doesn't directly bear on the objective.
If the goal is to give the kinetic energy of the ejection event some time to dissipate before the distance between the parts reaches the length of the shock cord, then the key variables are how much KE is there at the moment of separation and how rapidly will the parts dissipate that energy.
It's mentioned that a Cherokee-E is a longer rocket, so it should have a longer cord. However, just to make a point, I pulled up an OR sim of the Cherokee E. Reduced the overall length of the airframe by 50 percent and added 10 grams of clay to the tip of the nose cone to get stability back. Same basic rocket. Airframe volume pressurized by the ejection charge is less than half (because the pressurized volume excludes the volume occupied by the motor and mount). So the KE at the moment of separation should be considerably greater than with the unmodified model. Yet the rule of thumb relating to body length (or overall length, if one prefers that formulation) will have us using a much shorter shock cord.
In other words, if 10 ft. plus a leader the length of the body tube plus six inches is the right length of shock cord for the Cherokee E, it's probably also the right length, if not possibly too short, for, say, a Goblin.
Similar concept: The calculations for sizing BP ejection charges are discussed as targeting pressurizing the internal volume to be opened to ~15 psi. A longer rocket will need more powder, but so will a short/fat rocket of increased volume. The amount of BP to be used isn't just determined by looking at the length of the cavity.