The knowledge about future solar activity is necessary to plan our space-based missions. As yet, several prediction models (statistics-or dynamo-based) have been developed to forecast the peak smoothed sunspot number (SSN) of the upcoming solar cycle. Many of these data-based prediction models require sunspot number until the end of nth solar cycle to predict the peak of the solar cycle n + 1. However, one prefers to have the predictions well in advance. We propose a new data-based model that can provide information about the peak SSN (or amplitude) of solar cycle n + 1, and sum of peak SSN of solar cycle n + 2 and n + 3 at the end of each nth solar cycle. The solar cycles are paired using even-odd cycles, therefore in this model n is allowed to take even number. We used recently updated, Version-2, daily and monthly sunspot-number data. The area under the curve of each nth solar cycle [A(n)] is estimated and used in the model together with its length and peak SSN [S-max(n)]. We noticed that difference in the area under the curve of solar cycles n and n + 1 can be used to predict the summation of peak SSN of n + 2 and n + 3 solar cycles (i.e., S-max(n+2) + S-max(n+3)). We have tested this prediction model with both daily and monthly sunspot numbers. Our model predicts; S-max(24) + S-max(25) = 219.7 +/- 31 and S-max(26) + S-max(27) = 229.4 +/- 31. As the peak SSN of solar cycle 24 is known, we get S-max(25) = 103.3 +/- 15. Further, model suggest that the solar cycle 26 and 27 would be similar or slightly stronger than solar cycle 24 and 25.