Matching empirical impulse responses (Gali's recommendation)

In Chapter 3 of Gali (2015), he mentions “…matching some of the quantitative features of the empirical impulse responses requires that the basic New Keynesian model be enriched in a variety of dimensions”.

"variety of dimensions" seems very broad though and Gali does not expatriate. So my guess is either more country-specific features and also perhaps adjusting parameter values?

My problem
I have a model I want to use for policy analysis for a specific economy. I have some empirical results from SVAR models I want to incorporate into the theoretical model before the policy analysis part. I have incorporated some country-specific features in the model but that in itself does not replicate the empirical responses I want to incorporate. After adjusting some parameter values, the matching gets better than before. So I now want to use the model for policy analysis. Does my approach here make sense, regarding the incorporation of stylized facts via adjusting parameter values? And if I ignore this so that the theoretical IRFs do not match the empirical ones, can I still use the model for policy analysis?

I am trying to follow the following research program described in the book written by Favero, C. A. (2001). 1. Document stylized facts 2. Incorporate the stylized facts into the theoretical model. 3. Conduct policy analysis with the model. I think it is a similar concept in Gali.

The quote refers to e.g. the whole investment/capital part missing from the model.

Generally, your model needs to have all the features required to explain the facts at hand and needs to be appropriately calibrated. Generally, you cannot take parameters from a different country.

If model and VAR IRFs differ significantly, you need to investigate why that is the case. It may be a problem with the VAR like identification or your model is missing some important ingredient.